Using AeroNet data to parametrize atmospheric correction in ARCSI

[Anders Knudby]

Hi all, I would like to directly import an aeronet file to parametrize atmospheric correction using ARCSI, as is documented for Py6S here: http://py6s.readthedocs.org/en/latest/helpers.html#importing-aerosol-data-from-aeronet-data. I realize ARCSI is built on Py6S, but not sure whether this functionality exists, or whether I need to manually derive the -aot parameter (and make an educated guess at the -aeropro parameter as well) from the AeroNet data, and then put them into my ARCSI run. Any suggestions? Has anybody else done this?

A related question: For 'my' AeroNet station, I get AOT_500 values (as well as values at 340, 380, 440, 675, 870, 1020 and 1640 nm), but Py6S requires AOT_550 values, so presumably ARCSI also uses AOT_550 values? Are there 'standard' - as in documented/published - conversions I should be aware of before I go create an empirical fit (which seems to provide a very reasonable value)?

Anders

[Pete Bunting [pfb]]

Hi Anders, 

Sorry for my slow response. No I haven’t implemented that option - which is good idea. I’ll add it to the ‘to do’ list. 

Yes, ARCSI requires AOT at 550nm. Converting AOT values is not something I have needed to do so I am not well qualified to comment but an empirical fit of the AOT values you have done would seem appropriate. 

Best wishes, 

Pete

****************************************************
* Dr Pete Bunting
* Senior Lecturer in Remote Sensing
* Earth Observation and Ecosystem Dynamics Group
* Department of Geography and Earth Sciences
* Aberystwyth University
* Aberystwyth
* Ceredigion
* SY23 3DB
* UK
* 
* Ph: +44 (0) 1970 622615
* Mob: +44 (0) 7917 842743
* Email: p...@aber.ac.uk
* ORCID: http://orcid.org/0000-0002-7435-0148
****************************************************

--
You received this message because you are subscribed to the Google Groups "RSGISLib Support" group.
To unsubscribe from this group and stop receiving emails from it, send an email to rsgislib-suppo...@googlegroups.com.
To post to this group, send email to rsgislib...@googlegroups.com.
Visit this group at http://groups.google.com/group/rsgislib-support.
For more options, visit https://groups.google.com/d/optout.

[Dr Pete Bunting]

Hi Anders, 

It is in one of the sensor classes (e.g., arcsisensorlandsat7.py) in one of the functions: 

convertImageToSurfaceReflSglParam

convertImageToSurfaceReflDEMElevLUT

convertImageToSurfaceReflAOTDEMElevLUT

Are you just using a single value of AOT for the the correction then it will be convertImageToSurfaceReflSglParam. If you provide a single value of AOT but a DEM then it will be convertImageToSurfaceReflDEMElevLUT and if an AOT surface is calculated / inputted and the DEM is provided then it will be convertImageToSurfaceReflAOTDEMElevLUT.

At the end of the function there will be a call to RSGISLib 

convertImageToSurfaceReflSglParam:

rsgislib.imagecalibration.apply6SCoeffSingleParam

convertImageToSurfaceReflDEMElevLUT:

rsgislib.imagecalibration.apply6SCoeffElevLUTParam

convertImageToSurfaceReflAOTDEMElevLUT:

rsgislib.imagecalibration.apply6SCoeffElevAOTLUTParam

You can see the documentation for those functions here http://www.rsgislib.org/rsgislib_calibration.html

You have two options, the first is to change the scale value passed to the rsgislib function (i.e., 1000) or change the output data type to be float (list of datatypes here: http://www.rsgislib.org/rsgislib.html) so the values are not rounded. 

The reason I choose 1000 as the scale factor is that I didn’t think that the reflectance values would be meaningful beyond that level of precious but I am interested that you find that there is information in those regions. Maybe I should provide this as an option on via the terminal and environmental variables - I will add it to the list of things to do (https://bitbucket.org/petebunting/arcsi/issues) when I next get chance to work on ARCSI. 

Best wishes, 

Pete

****************************************************
* Dr Pete Bunting
* Senior Lecturer in Remote Sensing
* Earth Observation and Ecosystem Dynamics Group
* Department of Geography and Earth Sciences
* Aberystwyth University
* Aberystwyth
* Ceredigion
* SY23 3DB
* UK
* 
* Ph: +44 (0) 1970 622615
* Mob: +44 (0) 7917 842743
* Email: p...@aber.ac.uk
* ORCID: http://orcid.org/0000-0002-7435-0148
****************************************************

On 26 Jul 2015, at 21:10, Anders Knudby <knu...@gmail.com> wrote:

Hi Pete, very quick question. Could you point me to the code location of the multiplier used in ARCSI for surface reflectance values. I believe the one currently used is 1000 (i.e. a 1% reflectance pixel ends up with the value 10). I want to change it to 10000 because I work with water surfaces. I figured you could quickly tell me a) if an easy change is possible, and b) what file to look in to make the change.

Cheers, Anders

--

Anders Knudby, PhD
e-mail: knu...@gmail.com
skype ID: anders.knudby

Man's mind, once stretched by a new idea, never regains its original dimensions.
~ Oliver Wendell Holmes